Benjamin Reed commented on ZOOKEEPER-568:

i think that r.nextInt is in the right place. we should document why it is 
there. (so that all the replicas aren't snap shotting at the same time.)

apart from not counting the read only requests, we should also process them 
immediately if there are no pending writes.

> SyncRequestProcessor snapping too frequently - counts non-log events as log 
> events
> ----------------------------------------------------------------------------------
>                 Key: ZOOKEEPER-568
>                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-568
>             Project: Zookeeper
>          Issue Type: Bug
>    Affects Versions: 3.2.1
>            Reporter: Patrick Hunt
>             Fix For: 3.3.0
> Noticed the following issues in SyncRequestProcessor
> 1) logCount is incremented even for non-log events (say getData)
> txnlog should return indication if request was logged or not (if hdr ==null 
> it returns)
> also:
> 2) move r.nextInt below logCount++ (ie if an actual log event)
> 3) fix indentation after txnlog.append (for some reason has unnecessary 4 
> char indent)

This message is automatically generated by JIRA.
You can reply to this email to add a comment to the issue online.

Reply via email to